I wonder if clients could automatically calculate a spam score for relays and share this score with the network. Maybe if a user marks a note as spam it could implicitly add this to the relays spam score for the user.

once we have spam scoring algos in clients, we could derive a score for the relay automatically. This would give more insight for users, and for clients to recommend relays in manual mode. Autopilot mode (outbox) could use these scores for choosing which relays to pull from.

This is all annoying stuff i would rather not have to deal with, but it’s clear we will have to deal with it sooner than later.

The saddest part is that the bar for creating a usable nostr client is rising. I’m hoping to add as much of this logic into nostrdb as possible so native client devs don’t have to deal with all of this stuff in the future.

Reply to this note

Please Login to reply.

Discussion

DVMs

How would this work?

You subscribe to one or more DVMs that scores users. Then you just require their wot to show things on screen.

So much attention paid to spam which is actually an easy problem to solve (Dynamic sat per note and reply requirement and poof 💥 spam gets zapped as the sats per note/reply goes up). You are avoiding solving the real problem which is Child Pornography and just saying "It's part of reality, get over it". nostr:nprofile1qqsr9cvzwc652r4m83d86ykplrnm9dg5gwdvzzn8ameanlvut35wy3gpp4mhxue69uhkummn9ekx7mqprfmhxue69uhhg6r9vehhyetnwshxummnw3erztnrdaksz9rhwden5te0wfjkccte9ejxzmt4wvhxjmcgek0h3 the gaslighting scammer. CyberBeggar. nostr:nprofile1qqsgydql3q4ka27d9wnlrmus4tvkrnc8ftc4h8h5fgyln54gl0a7dgspzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqg5waehxw309aex2mrp0yhxgctdw4eju6t0qy2hwumn8ghj7un9d3shjtnddaehgu3wwp6kyfehcpn READ THIS!!!!! You've violated the ABC of life "Avoid Bad Company" these devs just want your money. Money grubbing hucksters.

I like the thinking. But it would also be prone to Sybil attacks.

May web of trusts is the actual response to both?

Yeah wot always needs to be involved

只是 Damus 中的垃圾比较多。 #nostr 并不是垃圾收容协议,它并不拒审查——人人可审。

You make it a prediction market where people can bet on what is spam and what is not.

I’m not sure this solves the problem at all. Isn’t that just forcing clients to switch from relay to relay that doesn’t yet have spam?

Let’s use WoT and for whoever isn’t in it, you need some minimal PoW if you want your post seen… spammers will need to spend a lot of energy for their many posts but for a single user the energy will be minimal.

Don’t want to spend cpu cycles? zap the OP and your replies become visible. 🤷

Its more just to provide information so when they want to switch to manual mode with a small relay set they know what relays to use

wot is the best foot forward. it solves a lot of issues and gives the viewer an informed opinion. all clients should implement this.

Are there proof of work relays?

What is a reasonable pow number to have that is also annoying for spammers? 10?

I like the idea of pow posting as it means you won't miss new people or people way outside your bubble(s)

powrelay.xyz

labour.fiatjaf.com

every time you increase the required number of work bits by one, say from 9 -> 10, you effectively double the amount of work for a spammer.

idk what would be a good requirement to set, but my goal would be to ensure replyguy would need to be running his script on a $50/mo server to keep going 😂

(Yeah I only said 10 cos you mentioned 10 lol)

I don't know how this would work tho. If I add these relays to will they automatically make it my notes are pow?

Nah, they require client support. Useless unless you build your own client.

I wish I could use those relays, but the popular Nostr clients don’t support PoW

We are stuck

zaps already support comments. Clients should show these comments accordingly as normal replies and allow replying. Then you could limit replies from people outside your WoT to those paying zaps, only.

what if someone spins up npubs and uses them to spam report notes as spam? ReportGuy?

Will be WoT based of course

Don’t see how spam scores will work. Eventually all your relays, except WoT ones will be blocked.

The fundamental problem is that there is no difference between a spammer and a new user. LLMs will make them indistinguishable eventually.

What other solutions are there except to impose a significant onboarding or ongoing cost to new users? There could be many ways to impose this cost but it seems we have to do it.

Spam filters, pattern matching, manual blocking is all a waste of dev resources because attackers will just escalate over time.

nostr:note1vgvhwjn6ewsr7z4cq0qc4j4fpz0aqrc66u5xt7r7fat35p5s09uqlvggjv

My mind immediately goes to paid relais gaming that system

they connect all the way up your ass and cause you to move and speak because you're a goddamn soulless robot

?name=image.png

Mean people suck. Nice people swallow. 🌹

So, was there a time that one guy was on his back while you rode him and either sucked the other guy or the other did anal? If I were on my back, it might be weird to view the other guy’s nutsack or, if anal, have his balls slap mine. I’d much rather stand.